Audio summarization of meetings driven by user participation

Patent metadata

Field	Value
Publication number	US-10204158-B2
Application number	US-201615076820-A
Country	US
Kind code	B2
Filing date	Mar 22, 2016
Priority date	Mar 22, 2016
Publication date	Feb 12, 2019
Grant date	Feb 12, 2019

Abstract

A method for recording and summarizing a conference is provided. The method enables a conference leader to record and produce a detailed summary of the conference. The detailed summary contains hot links to various portions and/or segments of what was discussed in the conference. This summary also enables team members who are not able to attend the conference, or those who need a quick review of the topics covered at the conference, to survey key topics covered, and have the option to replay selected portions of the conference.

First claim

What is claimed is:

1. A computer-implemented method comprising:
monitoring a conference for a set of phrases to be spoken into a microphone by a conference member, the set of phrases having respectively corresponding triggers defined before the conference occurs;
recording to an audio file a first audio segment of the conference for playback as a recorded discussion, the start of the first audio segment being identified by a first phrase of the set of phrases;
responsive to detecting a second phrase of the set of phrases, stopping the recording of the first audio segment, and recording to the audio file a second audio segment of the conference, the second phrase being a transitional phrase;
generating, by a processor, a first keyword corresponding to the first audio segment of the conference;
determining, by the processor, two keywords including the first keyword of the first audio segment and a second keyword of another audio segment of the audio file are related;
generating a selectable list of audio segments recorded to the audio file including the first audio segment and the second audio segment, the selectable list including segment links to the audio segments of the audio file;
generating a relationship link in the selectable list for the first audio segment and the second audio segment based on the two keywords being related;
identifying, by the processor, a set of related audio segments in the audio file based, at least in part, on a context of the related audio segments;
generating a listing of the set of related audio segments including a start time corresponding to each of the related audio segments, the start time being with respect to a playback duration of each related audio segment;
storing, to a storage device, a summary of the conference including the audio file, the selectable list of audio segments, and a listing of keywords corresponding to audio segments in the audio file including the first keyword and the second keyword; and
converting, by the processor, the summary of the conference into a text transcript including the listing of keywords and a corresponding location of the audio segments.

2. The method of claim 1 , wherein the text transcript includes a selectable link to a location of an audio segment.

3. A computer program product comprising a non-transitory computer readable storage medium having a set of instructions stored therein which, when executed by a processor, causes the processor to summarize a conference by:
monitoring a conference for a set of phrases to be spoken by a conference member, the set of phrases having respectively corresponding triggers defined before the conference occurs;
recording to an audio file a first audio segment of the conference for playback as a recorded discussion, the start of the first audio segment being identified by a first phrase of the set of phrases;
responsive to detecting a second phrase of the set of phrases, stopping the recording of the first audio segment, and recording to the audio file a second audio segment of the conference, the second phrase being a transitional phrase;
generating a first keyword corresponding to the first audio segment of the conference;
determining two keywords including the first keyword of the first audio segment and a second keyword of another audio segment of the audio file are related;
generating a selectable list of audio segments recorded to the audio file including the first audio segment and the second audio segment, the selectable list including segment links to the audio segments of the audio file;
generating a relationship link in the selectable list for the first audio segment and the second audio segment based on the two keywords being related;
identifying a set of related audio segments in the audio file based, at least in part, on a context of the related audio segments;
generating a listing of the set of related audio segments including a start time corresponding to each of the related audio segments, the start time being with respect to a playback duration of each related audio segment;
storing a summary of the conference including the audio file, the selectable list of audio segments, and a listing of keywords corresponding to audio segments in the audio file including the first keyword and the second keyword; and
converting the summary of the conference into a text transcript including the listing of keywords and a corresponding location of the audio segments.

4. The computer program product of claim 3 , wherein the text transcript includes a selectable link to a location of an audio segment.

5. A computer system comprising:
a processor set; and
a computer readable storage medium;
wherein:
the processor set is structured, located, connected, and/or programmed to run program instructions stored on the computer readable storage medium; and
the program instructions which, when executed by the processor set, cause the processor set to summarize a conference by:
monitoring a conference for a set of phrases to be spoken by a conference member, the set of phrases having respectively corresponding triggers defined before the conference occurs;
recording to an audio file a first audio segment of the conference for playback as a recorded discussion, the start of the first audio segment being identified by a first phrase of the set of phrases;
responsive to detecting a second phrase of the set of phrases, stopping the recording of the first audio segment, and recording to the audio file a second audio segment of the conference, the second phrase being a transitional phrase;
generating a first keyword corresponding to the first audio segment of the conference;
determining two keywords including the first keyword of the first audio segment and a second keyword of another audio segment of the audio file are related;
generating a selectable list of audio segments recorded to the audio file including the first audio segment and the second audio segment, the selectable list including segment links to the audio segments of the audio file;
generating a relationship link in the selectable list for the first audio segment and the second audio segment based on the two keywords being related;
identifying a set of related audio segments in the audio file based, at least in part, on a context of the related audio segments;
generating a listing of the set of related audio segments including a start time corresponding to each of the related audio segments, the start time being with respect to a playback duration of each related audio segment;
storing a summary of the conference including the audio file, the selectable list of audio segments, and a listing of keywords corresponding to audio segments in the audio file including the first keyword and the second keyword; and
converting the summary of the conference into a text transcript including the listing of keywords and a corresponding location of the audio segments.

6. The computer system of claim 5 , wherein the text transcript includes a selectable link to a location of an audio segment.

Assignees

• IBM

Inventors

• Hay Alexander T
• Imbamba Eugene K
• White Terrence E

Classifications

• G06F17/30772

  Physics · mapped topic

• G06F17/30743

  Physics · mapped topic

• G10L15/1822

  Parsing for meaning understanding · CPC title

• H04L65/403

  Arrangements for multi-party communication, e.g. for conferences (data switching systems for conference H04L12/18; arrangements for connecting several subscribers to a common circuit, i.e. affording conference facilities H04M3/56; television conferencing systems H04N7/15) · CPC title

• G06F17/30778 Primary

  Physics · mapped topic